Academic journal article Memory & Cognition

Learning Categories by Making Predictions: An Investigation of Indirect Category Learning

Academic journal article Memory & Cognition

Learning Categories by Making Predictions: An Investigation of Indirect Category Learning

Article excerpt

Categories are learned in many ways, but studies of category learning have generally focused on classification learning. This focus may limit the understanding of categorization processes. Two experiments were conducted in which participants learned categories of animals by predicting how much food each animal would eat. We refer to this as indirect category learning, because the task and the feedback were not directly related to category membership, yet category learning was necessary for good performance in the task. In the first experiment, we compared the performance of participants who learned the categories indirectly with the performance of participants who first learned to classify the objects. In the second experiment, we replicated the basic findings and examined attention to different features during the learning task. In both experiments, participants who learned in the prediction-only condition displayed a broader distribution of attention than participants who learned in the classification-and-prediction condition did. Some participants in the prediction-only group learned the family resemblance structure of the categories, even when a perfect criterial attribute was present. In contrast, participants who first learned to classify the objects tended to learn the criterial attribute.

Learning about categories and making categorical decisions are important for intelligent behavior. Knowledge about an object's category membership provides information about how that object should be used, manipulated, approached, or avoided. Psychologists have often studied categorization by training participants in a classificationlearning task in which the participants classify exemplars one at a time, with feedback on every trial. In this article, we argue that a focus on classification learning can lead to a restricted view of category learning. Although other means of category learning have been examined recently, much still remains to be learned about how different ways of learning might affect the category representation. The goal of this article is to explore this issue by introducing an indirect way of learning categories that we believe is common, provides a clear contrast to classification learning, and allows a better understanding of the theoretical implications of nonclassification category learning.

Classification Learning

Research in which the classification paradigm has been used has supported several different theories of categorization, including the idea that people represent categories as prototypes (Blair & Homa, 2001; Minda & Smith, 2001; E. E. Smith & Medin, 1981; J. D. Smith & Minda, 1998), as collections of exemplars (Medin & Schaffer, 1978; Nosofsky, 1987; Nosofsky & Zaki, 1998), and as rules and exceptions (Nosofsky, Palmeri, & McKinley, 1994; Palmeri & Nosofsky, 1995; E. E. Smith, Patalano, & Jonides, 1998). These theories emphasize categories as something that participants use to classify a new item, and they emphasize the importance of a contrast category in making a decision. In other words, in order to be useful for making classification decisions, two prototypes should be distinguishable from each other or two clusters of exemplars should be distinguishable from each other. Participants learn to attend to the features that maximize this distinction (Kruschke, 1992; Nosofsky, 1987).

Although these properties are important in categorization, it is possible that their primary importance in these theories comes from a reliance on the classificationlearning paradigm. Classification explicitly requires participants to distinguish one family of exemplars from another and may affect how a category is represented. In situations outside the lab, people may not interact with an item solely by classifying it. Rather, people may use classification as a step in attaining some goal for which knowledge of an object's category is useful. A thorough examination of category learning must include learning tasks that require participants to do more than simply classify an item and must specify how different learning environments shape category representations. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.